VMware on Tuesday announced updates to its server virtualization software vSphere, including the latest progress in its partnership with Nvidia aimed at accelerating AI adoption. The company also announced updates to its storage virtualization portfolio vSAN 7 and said that both of the new releases will support enterprises with developer- and AI-ready infrastructure.
On the server virtualization front, VMware announced that Nvidia AI workloads are certified to run on vSphere 7. The new Nvidia software suite, officially named Nvidia AI Enterprise, aims to give enterprises the ability to develop AI software for a range of applications such as advanced diagnostics in healthcare, smart factories for manufacturing, and fraud detection in financial services.
According to VMware and Nvidia, the software suite provides multi-node, AI application performance on vSphere that is indistinguishable from bare-metal servers. Currently, vSphere 7 is the only virtualization platform certified to support the Nvidia AI software suite.
"Until now, the world has run AI on bare-metal servers," said Justin Boitano, VP and GM of Enterprise and Edge Computing at Nvidia. "Nvidia AI Enterprise enables customers to reduce AI model development time from 80 weeks to just eight weeks, and allows them to deploy and manage advanced AI applications on VMware vSphere with the same scale-out, record-breaking NVIDIA accelerated computing performance that's possible on bare metal."
To power the applications in Nvidia AI Enterprise, VMware said vSphere 7 Update 2 is certified for Nvidia A100 Tensor Core GPUs on NVIDIA-certified Systems, including those from Dell Technologies, HPE, Lenovo and Supermicro. Nvidia has also certified vSphere as the only server virtualization software to provide hypervisor support for live migration with Nvidia multi-instance GPU technology.
The VMware vSphere 7 Update 2 and VMware vSAN 7 Update 2 are available today. Nvidia AI Enterprise is available as a perpetual license at $3,595 per CPU socket. Enterprise Business Standard Support for the suite is $899 annually per license.
In September, VMware and Nvidia unveiled plans to integrate the latter's artificial intelligence applications for unified management of apps, security and data processing unit accelerators. The partnership secured Nvidia's role in hybrid clouds as VMware outlined an architecture incorporating data processing units (DPUs) in the data center, cloud and edge. Both companies said the bet was that the integration would be able to speed up AI adoption in enterprises.